ERIC EJ905715: Who Benefits from Tuition Discounts at Public Universities?

2010·0.11 MB·English

by  ERIC

Who Benefits from Tuition Discounts
at Public Universities?
By Nicholas W. Hillman
Nicholas W. Hillman is This article uses data from the 2004 National Postsecondary Student Aid Study to
assistant professor of provide insight about the range of tuition discounting practices at public
educational leadership and institutions. Specifically, it examines the characteristics of students who receive
policy studies at the tuition discounts from public four-year colleges and universities. A binary logistic
University of Utah. regression is applied to all students enrolled in a public four-year institution in
2003-04 to determine which students were most likely to receive a tuition discount.
Descriptive statistics show how tuition discount rates differ according to student
characteristics such as academic level, race, residency, family income, and
institution type. This study examines national patterns and also makes use of the
12-state representative samples available in NPSAS:04. Findings demonstrate that
low-income students, minorities, freshmen, and non-resident students are the
groups most likely to receive tuition discounts. Based on the logistic regression
results, it appears that public colleges are using tuition discounts in a way that
increases opportunity for traditionally underrepresented students, and as an
incentive for freshmen to enroll in college. However, while low-income students are
more likely to receive tuition discounts, their discount rate is equal to or less than
their upper-income peers.
S
tate and federal government agencies have traditionally taken the lead in
providing grant aid to students attending public colleges and universities.
Since state and federal expenditures for higher education have failed to
keep pace with rising tuition costs and student enrollment levels (Toutkoushian,
2001; Weerts & Ronca, 2006), public colleges have provided students with
grants from their own funds. This was not the case 20 years ago, when it was
less common for public colleges to provide institutional grants to students
(Heller, 2000). Given the tenuous support from state and federal sources, public
colleges have taken it upon themselves to become an additional source of aid
for students, but not much is known about the ways in which public colleges
spend institutional aid (College Board, 2006).
This paper utilizes data from the 2004 National Postsecondary Student Aid
Study (NPSAS:04), to examine the characteristics of students who receive
tuition discounts from public four-year colleges and universities. This analysis
selects all students who were enrolled at public four-year institutions in 2003-04
and applies a binary logistic regression to determine which of those students
were most likely to receive a tuition discount. Additionally, it utilizes descriptive
statistics such as means, percentages, and ranges to show how tuition discount
rates differ based upon various student characteristics. This study provides
information on national patterns, but it also makes use of the 12-state
representative samples available in NPSAS:04. Enrollment management
personnel, financial aid directors, higher education policy researchers, and
campus planning officials will find these results informative as this paper
provides insight about the range of tuition discounting practices at public
institutions.
National Association of Student Financial Aid Administrators 17
Review of the The majority of tuition discounting research focuses on private four-year
Literature institutions; few studies examine the practice at public colleges and universities
(Redd, 2000; Lapovsky & Hubbell, 2001; College Board, 2006). Tuition
discounting has a long history at private institutions, dating back to the 19th
century when some colleges offered remissions to help “worthy” students pay
for college (Wilkinson, 2005). Private colleges rely heavily on tuition revenue as
a primary source of income, so discounting strategies are more common and
have a longer track record at private institutions. Since tuition discounting at
public institutions is a practice with a relatively short history (Heller, 2000), the
body of research on the range of practices is not extensive.
Over the past several years, state and federal spending on public higher
education has not kept pace with the rising costs and increased number of
students enrolling (Toutkoushian, 2001, Weerts & Ronca, 2006), causing public
colleges to become increasingly reliant on other sources of revenue including
tuition. During this same period, federal and state need-based financial aid
programs have been placed in a precarious position. The purchasing power of the
Federal Pell Grant has steadily declined since the 1980’s (College Board, 2007; St.
John, 2005) and states have invested heavily in merit-based (rather than need-
based) state grant programs (NASSGAP, 2006; Heller, 2002). These external
circumstances have put pressure on public institutions to use tuition discounts as
a way to help leverage aid for students with financial need. Institutional grant aid
is often considered the “financial aid of last resort” (Allan, 1999a), so public
institutions play an immensely important role in leveraging aid to students who
need it most in order to help them succeed in college.
In addition to providing need-based aid, institutions use merit-based
discounts as an enrollment strategy to recruit a desired mix of students based
on academic or athletic talent, residency status, or race and ethnicity (Allan,
1999b; Hossler, 2000; DesJardins & Bell, 2006). As colleges compete for talented
students to increase prestige and college rankings, they must weigh the
opportunity costs and ethics of choosing merit over need.
Institutional aid, whether it is need-based or merit-based, plays an important
role in college choice and student success. Students choose to attend college
based on iterative steps of information-gathering, and after weighing the costs
and benefits, will enroll in the institution that they perceive as the best fit
(Hossler, Schmit, & Vesper, 1999). Tuition discounting, therefore, becomes a
very powerful recruitment tool, especially for price-sensitive students who
might not enroll if the “sticker price” is beyond their ability to pay. Studies have
found financial aid to have a significant impact on student enrollment decisions
(Heller, 1997; St. John, 1996). Students who receive scholarships and grants
increase their likelihood of staying in college, so it behooves public institutions
to utilize tuition discounts in ways that maximize students’ likelihood of
continuing their enrollment.
During the late 1990’s and early 2000’s, several studies and national reports
focused on tuition discounting practices (Lee & Clery, 1997; Allan, 1999; Heller,
2000; Redd, 2000; Lapovsky & Hubbell, 2001, Davis, 2003), but there has been
little research on the subject in recent years. In 2006, The College Board
published a report titled Tuition Discounting: Not Just a Private Practice,
signaling a renewed interest in the topic. Not only did this report bring tuition
discounting back on the research agenda, it specifically drew attention to the
need for gaining a better understanding of the practice at public institutions.
18 Journal of Student Financial Aid Volume 40 • Number 1 • 2010
Data & Methods This study selects a national sample (n=31,542) of undergraduate students from
the 2004 National Postsecondary Student Aid Study (NPSAS:04) and describes
which students are most likely to receive tuition discounts from public four-year
colleges and universities. It also identifies the differences in tuition discount
rates depending on certain student characteristics. My primary research
questions are as follows: At public four-year colleges and universities, to what
extent do tuition discount rates differ based upon student characteristics such as
socio-economic background, college experience, and college choice?
Additionally, which students are most likely to receive discounts from public
four-year institutions?
Variables were selected based on their ability to serve as proxies for socio-
economic background, college experience, and college choice. St. John’s
workable models approach (1992) advocates for analyzing student-level data
that links sociological and economic data to higher education theory and
research. Therefore, I ensured that student socio-economic background factors
such as family income level gender, and parents’ level of education were
included in the analysis. Studies on college choice (St. John, Paulson, & Starkey,
1996; Hossler, Braxton, & Coopersmith, 1989) indicate that proximity and
institutional type, along with high school preparation, are significant factors for
influencing enrollment decisions. For these reasons, this study included
students’ residency status, Carnegie classification of their college, and type of
high school as factors that influence college choice. Student persistence
research (Tinto, 1993; Bean & Eaton, 2000) and studies on enrollment
management (Hossler & Bean, 1990; Hossler, 2000) indicate that academic
integration is a significant predictor of student success, so the study uses college
grade point average, choice of major, dependency status, and enrollment status
as factors relevant to the use of institutional aid in enrollment management.
For each variable, the average discount rate is calculated as described in the
following formula (College Board, 2006):
Average Institutional Grant Per Student
Average Tuition and Fee
The terms “tuition discount” and “institutional grant” are used
interchangeably in this article because they are both measuring the same
outcome. For example, a 15 percent discount rate means that for every $1,000
charged in tuition, the institution provides $150 in institutional grant aid.
In addition to the calculating students’ average discount rates descriptive
statistics are provided (see Table A) of awards to all students including the
percentage of students receiving institutional grants. These descriptive statistics
provide information about national discounting trends, but NPSAS:04 also
includes state-level data in which 12 states (California, Connecticut, Delaware,
Georgia, Illinois, Indiana, Minnesota, Nebraska, New York, Oregon, Tennessee,
and Texas) have robust representative samples. These descriptive statistics will
answer my first research question about the extent to which tuition discount
rates differ depending on student characteristics.
In order to answer the second research question about the likelihood of
receiving discounts at public four-year institutions, I chose to utilize binary
logistic regression where the dependent variable is whether a student received a
tuition discount (1) or did not receive a discount (0). Logistic regression is an
National Association of Student Financial Aid Administrators 19
appropriate technique to use when studying categorical outcomes (Long, 1997;
Peng, So, Stage, St John, 2002), whereas the binary logistic regression formula is:
ln( P )= ∝+ßx+ßx+…ßn
1-P 1 2 1 2 k k
In this formula, Pis the probability of a student receiving a tuition discount,
ln( P )is the natural logarithm of the odds of P, ∝is the Y intercept, ßis the
1-P
slope parameter, and x represents each predictor variable.
k
The logistic regression formula provides odds ratios for each predictor
variable. For example, when a predictor variable’s odds ratio is greater than
one, then groups of individuals represented by that variable are more likely
(than the reference group) to receive a tuition discount. If a predictor variable’s
odds ratio is less than one, then the opposite is true; when a predictor variable’s
odds ratio is less than one, then groups are less likely (than the reference group)
to receive a tuition discount.
Results In 2003-04, the national average discount rate for students attending public four-
year colleges and universities in the U.S. was 14.3 percent. This finding
corresponds with a recent analysis by the College Board (2006) that estimated
the average discount rate to be 15.3 percent when using a different database
that used institutions as the unit of measure. Approximately 1.19 million
students received tuition discounts from public four-year colleges in 2003-04,
accounting for 20 percent of the student body. At private four-year colleges,
approximately 1.27 million students, or 50 percent of the private sector student
body, received discounts in 2003-04. It is evident that private institutions engage
in discounting practices to a greater extent than the public sector, but in raw
number of students (1.19 and 1.27 million respectively), there is not much
difference in the total number of students receiving discounts. When examining
the three primary sources of student grant aid (federal, state, and institutional),
tuition discounts at public four-year colleges accounted for $3.5 billion in 2003-
04, which is $1.1 billion more than the total amount provided by state grant
programs. Federal grant aid accounted for approximately $4.3 billion, showing
that students benefit most from federal grants, then institutional grants, and
lastly state grants.
Considering the scope of the tuition discounting practice at public four-year
colleges, researchers and policymakers do not have much data about the
students who receive these awards. Tuition and fees is one of the fastest-
growing sources of revenue for public institutions, and that access and
affordability are among the most important higher education policy
discussions, making it critical for these stakeholders to know who is benefitting
the most from redistributing funds via discounts.
Class Level.Most institutional aid is awarded to students early in their academic
careers, rather than later; 23.4 percent of freshmen received tuition discounts,
while 21.1 percent of seniors and 15.3 percent of fifth-year seniors received
tuition discounts. Using the logistic regression model, it is evident that as
students progress through public four-year colleges, they are significantly less
likely to receive tuition discounts. Seniors, junior, and sophomores all have
statistically significant (α<.001) odds-ratios that are less than one, indicating
that they are less likely than freshmen to receive tuition discounts. This
confirms the descriptive statistics that colleges are front-loading tuition
20 Journal of Student Financial Aid Volume 40 • Number 1 • 2010
discounts to freshmen and are less likely to award discounts to students as they
progress in college.
Freshmen aren’t only more likely to receive discounts than other students, but
their average discount rate is higher than that of any other class. The average
discount rate for freshmen is 16.2 percent discount, while seniors receive a 13.4
percent discount and fifth-year seniors receive a 7.9 percent discount rate.
Residency Status.Approximately 30.4 percent of non-resident students receive
tuition discounts at public four-year colleges, as opposed to the 19.7 percent of
in-state residents. Students who go out-of-state to attend college are
significantly more likely to receive tuition discounts than their in-state peers. As
the regression results indicate, non-residents have odds of receiving tuition
discounts that are1.717 times higher than in-state. While non-resident are more
likely to receive discounts, it is also important to note that they receive greater
discount rates than in-state students (16.6 and 13.6 percent, respectively).
Although the difference in the average discount rate for residents and non-
residents is only three percentage points, the magnitude of that small difference
is large. For example, a one-percent increase to the non-resident tuition
discount rate would cost an institution $100.13, while an equivalent percentage
increase for resident students would only cost $35.80.
National discount rates are useful in setting a benchmark for institutions, but
it is important to note that these discount rates vary widely among the 12 states
in which NPSAS:04 provides representative samples (see Tables 1 and 2). Public
four-year colleges in states with large grant programs (New York, Georgia,
Indiana, Minnesota, and Tennessee) tend to provide considerably smaller
discounts than the national average. Fewer than 20 percent of in-state students
in these states receive tuition discounts, and their rates are considerably lower
than the national average.
Race. Minority students are more likely than White students to receive tuition
discounts at public four-year institutions. Approximately 19.8 percent of White
students receive tuition discounts, while Black, Asian, and Hispanic students are
the most likely to receive discounts (22.0, 26.0, and 23.3 percent respectively).
Students’ race/ethnicity is one characteristic where we see the widest range of
tuition discounting practices. The average discount rate ranges from a low of
12.3 percent to a high of 21.0 percent, which is among the largest ranges of all
variables in this study. Although Hispanic students are 1.2 times more likely
than White students to receive tuition discounts, their average discount rates
(12.6 percent) are smaller than White students’ rates (13.3 percent).
Family Income.As family adjusted gross income levels increase, the likelihood
of receiving tuition discounts decrease, holding all other variables constant.
When compared to middle-income families (AGI $50,000 - $70,000), students
whose income is below $30,000 are 1.896 times more likely to receive tuition
discounts. That likelihood ratio steadily decreases as we move higher in the
income categories; students from the highest income group (AGI greater than
$100,000) have an odds ratio of 0.571, which means they are nearly two times
less likely than middle-income students to receive tuition discounts.
Although low-income students are the most likely to receive tuition discounts
at public four-year institutions, their average discount rates are actually not
significantly different than their upper-income peers. Students whose families
earn between $70,000 and $100,000 receive an average discount of 15.1 percent,
while the lowest income students only receive a 14.7 percent discount rate.
National Association of Student Financial Aid Administrators 21
Middle- and upper-income students receive almost the same (and sometimes
even greater) discount rates as low-income students, indicating that colleges are
spending institutional funds on students who do not necessarily have financial
need. This trend is a significant departure from the original intent of
institutional aid (Redd, 2000), as colleges originally utilized discounts as a
means for helping low-income students cover unmet financial need.
In California, Minnesota, and Texas, public four-year colleges award greater
tuition discounts to lower income students than to higher income students. The
opposite story is true in Indiana, Nebraska, and Tennessee, where higher
income students receive greater discount rates than lower income students. For
example, low-income students in Texas receive a 15 percent tuition discount,
while high-income students only receive a 5 percent discount. In Nebraska,
however, the highest income students receive a 40 percent discount, while low-
income students only receive a 20 percent discount.
Institutional Type.It is important not to overlook the differences among
“types” of public four-year colleges and universities. Research-intensive and
doctoral degree granting institutions have different missions than regional and
open-access institutions or teaching colleges. Therefore, it is necessary to
differentiate discount rates for students attending different types of public four-
year institutions. Students attending master’s and baccalaureate degree granting
institutions are significantly less likely (odds ratio = 0.73) to receive tuition
discounts than those attending doctoral and research institutions. Students
attending doctoral and research universities receive average discounts of 16.3
percent, while students at master’s and baccalaureate institutions receive
average discounts of 11.3 percent.
Summary and This analysis provides a broad overview of tuition discounting practices at
Discussion public four-year colleges and universities and shares information about the
profile of students who receive these awards. This study found that low-income
students, minorities, freshmen, and non-resident students were the groups most
likely to receive tuition discounts. Based on the logistic regression results, it
appears that public colleges are utilizing tuition discounts in such a way that
increases opportunity for traditionally underrepresented students, and as an
incentive for freshmen to enroll in college.
However, when examining the discount rates for various types of students,
institutional commitment to increasing opportunity becomes less evident,
particularly for low-income and minority students. Although Hispanic students
are more likely to receive discounts from public colleges, their average discount
rate is considerably lower than the national average. A similar story holds true
with regard to low-income students; they are more likely to receive discounts,
but their average awards are not higher than other students’ awards.
While low-income students are more likely to receive tuition discounts, their
discount rate is equal to or less than their upper-income peers. In some states,
discounts received by low-income students are nearly half as large as those
received by their upper-income classmates, signaling an imbalance in social justice
and an inequitable distribution of public funds. Tuition discounting can be a
strong predictor of student retention and it has become a useful tool for helping
price-sensitive students succeed in college. However, when discounts are awarded
to students who already have a high ability to pay, institutions fail to capitalize on
the opportunity to maximize student success for needy students.
22 Journal of Student Financial Aid Volume 40 • Number 1 • 2010
Public universities are utilizing tuition discounts to balance the often-
competing institutional objectives of increasing opportunity while
simultaneously increasing prestige. They are pressured to fill the financial gap
left from the declining purchasing power of the Pell Grant and the tenuous
state support for funding, while at the same time they must compete with other
institutions in order to improve institutional prestige and reputation. Given that
context, tuition discounting practices straddle a delicate line that requires
financial aid administrators and planners to determine “who” should receive
the institution’s limited financial resources.
Additionally, further study should examine the complex relationship of grant
aid practices to understand the distributional relationship between federal,
state, and institutional grant aid. Georgia serves as a good example of how to
conceptualize such a study. Since its state grant program provides financial
support to many resident students, Georgia colleges seem to have an incentive
to discount tuition for non-residents, thus giving Georgia colleges greater
flexibility for crafting a class. This study does not provide enough detail about
the fiscal federalism of grant aid, but future studies of tuition discounting in the
public sector should address these intended and unintended influence of
external aid. This study was designed to be descriptive in nature as an attempt
to fill a gap in the body of research on tuition discounting practices. Not only is
there a need to gain a better understanding of tuition discounting at public
institutions, but future tuition discounting studies should continue to focus on
the student (as opposed to the institution) as the primary unit of measure. By
focusing on student characteristics, researchers will be able to identify how
institutional aid policies are impacting student behavior while simultaneously
considering the intended and unintended consequences of redistributing
institutional aid based on merit versus need.
National Association of Student Financial Aid Administrators 23
Resources
Allan, R. (1999a). Taxonomy of tuition discounting. Journal of Student Financial
Aid, 29(2), No. 2.
Allan, R. (1999b). Tuition discounts, institutional student and aid scholarship
allowance.Washington, DC: Sallie Mae Education Institute.
Bean, J. & Eaton, S. B. (2000). A psychological model of college student retention.
Reworking the Student Departure Puzzle: Vanderbilt University Press.
Bishop, J., Wobmann, L. (2004, April). Institutional effects in a simple model of
educational production.Education Economics, 12(1).
College Board, (2006). Tuition discounting: Not just a private college practice.The
College Board. New York, New York
College Board, (2007). Trends in college pricing 2007.The College Board. New
York, NY.
Davis, J. (2003) Unintended consequences of tuition discounting. Lumina
Foundation for Education New Agenda Series, 5(1).
DesJardins, S. L., Bell, A. (2006). Using economic concepts to inform enrollment
management.New Directions for Institutional Research, 132.
Ehrenberg, R. (2006). Crafting a class: the trade-off between merit scholarships and
enrolling lower-income students.The Review of Higher Education, 29(2), 195-211.
Eisenhardt, K. (1989). Agency theory: An assessment and review.The Academy of
Management Science Review, 14(1), 57-74.
Heller, D. (1997) Student price response in higher education: An update to Leslie and
Brinkman. The Journal of Higher Education, 68(6).
Heller, D. (2000). Institutional scholarship awards: The role of student and
institutional characteristics. Presented at the annual meeting of the Association
for the Study of Higher Education. Sacramento, CA, November 16-19, 2000.
Heller, D. (2002, August). Merit scholarship and college access: Evidence from Florida
and Michigan. Who should we help? The negative social consequences of merit aid
scholarships. Harvard Project on Civil Rights. Cambridge, MA.
Hossler, D. (2000). The role of financial aid in enrollment management. New
Directions for Student Services, No. 89, pp. 77-90.
Hossler, D., Bean, J. (1990). The strategic management of college enrollments. San
Francisco: Jossey-Bass Inc.
Hossler, D., Braxton, J., Coopersmith, G. (1989). Understanding student college
choice.In J. Smart (Ed.) Higher Education Handbook of Theory and Research.
New York: Agathon Press.
Lapovsky, L. (2001). Institutional financial health: tuition discounting and enrollment
management.Study of College Costs and Prices, Vol. 2: Commissioned papers.
The Institute for Higher Education Policy (ed.)
Lapovsky, L., Hubbell, L. (2001). An uncertain future. The Business Officer, 34(8).
Lee, J. B., Clery, S.B. (1997) Institutional aid 1992-93. Postsecondary education
descriptive analysis reports, National Center for Education Statistics, Washington,
DC.
24 Journal of Student Financial Aid Volume 40 • Number 1 • 2010
Martin. R. (2004). Tuition discounting without tears. Economics of Education
Review, 23.
National Association of State Scholarship and Grant Aid Programs (NASSGAP),
(2006).37th annual survey report on state-sponsored student financial aid: 2005-06
academic year. Retrieved February 2009 from
http://www.nassgap.org/viewrepository.aspx?categoryID=3#
Sappington, D.E. (1991). Incentives in principal-agent relationships. The Journal of
Economic Perspectives, 5(2), 45-66.
St. John, E., Paulsen, M. & Starkey, J. (1996). The nexus between college choice and
persistence. Research in Higher Education, 37, 175-220.
St. John, E. P., (2000). The impact of student aid on recruitment and retention: What
the research indicates.New Directions for Student Services, 89, 61-75.
St. John, E.P. (2005, January). Affordability of postsecondary education: equity and
adequacy across the 50 states. Prepared for: Renewing Our Schools, Securing Our
Future – A National Task Force on Public Education. Center for American
Progress and the Institute for America’s Future.
Tinto, V. (1993). Leaving college: Rethinking causes and links of student attrition.
Chicago: University of Chicago Press.
Toutkoushian, R. (2001). Trends in revenues and expenditures for public and private
higher education.In Paulson, M. and Smart, J. (eds) The Finance of Higher
Education: Theory, Research, Policy, and Practice. New York, NY.
Weerts, D.J., Ronca, J. (2006). Examining differences in state support for higher
education: a comparative study of state appropriations for research I universities.The
Journal of Higher Education, 77(6).
Wilkinson, R. (2005).Aiding students, buying students: Financial aid in America.
Vanderbilt University Press, Nashville, TN.
National Association of Student Financial Aid Administrators 25
Table 1: Descriptive statistics of all students attending public four-year colleges
Percent
receiving Average Institutional Institutional Total
Tuition and fees tuition institutional merit-only need-based discount
(NPSAS) discount grant grants grants rate
Race/Ethnicity:
White $4,244 19.8% $565 $305 $184 13.3%
Black or African American $4,244 22.0% $891 $472 $319 21.0%
Hispanic or Latino $4,244 23.3% $535 $141 $344 12.6%
Asian $4,244 26.0% $781 $228 $482 18.4%
Other or multiple $4,244 21.8% $521 $181 $225 12.3%
Adjusted Gross Income (AGI):
Less than $30,000 $4,244 23.8% $625 $262 $295 14.7%
$30,000 - $50,000 $4,244 20.8% $567 $246 $247 13.4%
$50,000 - $70,000 $4,244 19.6% $572 $328 $175 13.5%
$70,000 - $100,000 $4,244 19.9% $640 $367 $165 15.1%
Greater than $100,000 $4,244 15.0% $605 $358 $158 14.2%
Gender:
Male $4,244 19.8% $599 $307 $221 14.1%
Female $4,244 21.7% $613 $290 $240 14.4%
Parent’s highest education level:
No college $4,244 20.9% $521 $223 $242 12.3%
Associate's degree or some college $4,244 19.6% $518 $230 $218 12.2%
Bachelor's $4,244 20.5% $624 $316 $222 14.7%
Post-baccalaureate or professional $4,244 22.0% $772 $432 $237 18.2%
Type of high school attended:
No high school diploma $4,244 14.4% $259 $106 $134 6.1%
Public $4,244 21.0% $591 $290 $231 13.9%
Private $4,244 21.1% $677 $325 $234 16.0%
Attended a foreign high school $4,244 22.2% $1,310 $715 $376 30.9%
Residency status:
Resident $3,632 19.8% $495 $225 $217 13.6%
Non-resident $9,755 30.4% $1,617 $958 $356 16.6%
Type of institution (Carnegie):
Research & Doctoral $5,149 24.6% $841 $393 $352 16.3%
Master’s & Baccalaureate $3,420 17.3% $385 $212 $113 11.3%
26 Journal of Student Financial Aid Volume 40 • Number 1 • 2010